18psi

iTrader: (76)

Nothing wrong with it. Joe is just trying to see if we can make it better.

hustler

iTrader: (6)

That would detonate any NA car I've tuned...severely. Elastic timing belt?

hustler

iTrader: (6)

Cams timing and better air-fuel homogenity due to turbulence when air is sucked through the intake valves at higher rpm.

The cam issue became clear to me when swapping from the factory cam to the MSM cam in my track car. The same, but opposite effect, in my VVT engine.

Before you chastize me for continuing to post in this thread with my non-factory engines, these experiences I discussed above help demonstrate the affects of altering the variables in your discussion.

Suddenly, those archaic small block Chevy engines with their offset spark plug location and peculiarly shaped pistons make a remarkable amount of sense and provide the mechanical octane I've recently discovered...in 50-year old technology.

I should have been an engineer...god damn calculus.

Leafy

iTrader: (1)

Dynamic compression ratio.

Everything below is highly theoretical because whats actually happening in the combustion chamber is not very well published.

Fuel/air distribution. More dispersed the fuel is the faster the flame speed. At higher rpms air and fuel are moving faster, so they'll get more dispersed. Also, hotter mixtures burn more efficiently, at higher RPMS you should get more adiabatic heating from the compression stroke. I've got no ******* idea how big of a factor that is or if the heat actually goes into the air charge though.

*Edit,
Hustler, what is that piston from exactly? Looks a lot like a piston from a DI engine, certainly going to help with fuel dispersion.

Joe Perez

iTrader: (8)

From post #1:

hustler

iTrader: (6)

78 Chevelle.
I assumed trigger latency adjustments would reduce the numbers in the spark table, not increase them. I guess I don't understand the concept. I adjust for this immediately with CAS cars, never touch it with CKP/CMP cars. I also didn't think latency adjustments were necessary as long as the latency inducing conditions are the same at all times...which they are unless you change timing belts.

baron340

iTrader: (5)

Since this thread is so far crammed full of extremely useful information, do you have an example of a reasonably well tuned map of yours for comparison? Preferably one on a stock 1.6/1.8. I'm starting to realize my timing map is probably stupid retarded.

hustler

iTrader: (6)

Not one that I can share. Post yours and I'll tell you what I think.

vehicular

iTrader: (3)

Subbed for observation purposes.

baron340

iTrader: (5)

I'll pull it up when I get home from work today. But I'm in the "jacked map from close enough setup online, thought it ran well enough and left it alone" crowd. I understand the theory of spark tuning well enough, but I've never put much time into actually tuning my ignition map.

Joe Perez

iTrader: (8)

At the risk of again derailing the thread:No, no. This has nothing at all to do with the timing belt.

There are three basic delays in the system which affect the relationship between crank (or cam) position and the spark event. They are:

1: The electrical delay between the time the tooth physically passes across the sensor and the time the sensor's output changes state.

2: The electrical delay between the time the sensor's output changes state and the time the that signal reaches the CPU itself (mostly caused by the time-constant of the ECU's input filters).

3: The electrical delay between the time the CPU commands the spark output driver to go low and the time that the actual spark event happens at the plug.

All three of these delays are a constant all all operating conditions, regardless of RPM, MAP, Dwell, commanded advance, etc. They will be different for different cars / sensors / ECUs / ignition systems, but for any given car, if the end-to-end delay is 100us, then it will always be 100us all the time (give or take minor changes due to temperature.)


When we set this latency variable, we're telling the CPU what the total lag through the system is so that it can compensate in software, by advancing the spark outputs by a certain fixed number of microseconds all the time.

If the latency setting is lower than the actual latency, then the actual spark angle will tend to retard as RPM goes up (because latency, as a percentage of time elapsed between trigger and spark, increases as RPM increases). If the latency setting is too high, then the CPU will be over-compensating and the spark will tend to self-advance as RPM increases.


Adjusting this is simple. You lock the advance to a specified number in the software (say, 10°) which overrides whatever happens to be in the spark map. Strobe the engine to ensure that this is the advance which are are actually seeing at idle.

Next, raise the RPM to, say, 4000, and strobe the engine again. Is the timing mark still pointed at 10°? If so, your hardware latency is correct. If the timing advanced beyond 10°, you need to decrease the value of HW latency. If the timing became retarded below 10°, you need to increase the latency.

So again, this is purely an electrical / software thing. It has nothing to do with mechanical slop in the engine.

hornetball

iTrader: (4)

Cut out a bunch from Joe's great discussion. Trying to get to the gist.

The main factor here is that the VE of the engine is changing as you go through the RPMs. What is VE really? It's how much fuel and air you can get crammed into your cylinder on the intake stroke. This in turn gets converted to pressure within the cylinder during combustion, which in turn changes the flame speed.

Typically, VE will be low at low RPM, increasing to some maximum for which the engine's valve timing has been tuned, and then tapering off as you reach your RPM limit. You can see this in the tuned advance curves where the advance doesn't increase as much as you might expect in the heart of the torque-producing RPMs (the plateau that Joe's questioning), but then you do see advance at the end where the engine "runs out of breath."

So many variables. Makes my head hurt.

triple88a

iTrader: (2)

You're missing the afr change. Closer to 11.5-12afr the flame front is much faster than at 16 afr.

triple88a

iTrader: (2)

Judging by those graphs

Joe Perez

iTrader: (8)

So, this is predicated on the assumption that I am linearly increasing AFR from 14.7:1 towards 12:1 as my RPMs increase beyond the point at which the advance curve flattens out.

That's fair.

Braineack

iTrader: (62)

Joe did you see the oem AFR map I posted? i feel like everything I post in here is ignored.

Joe Perez

iTrader: (8)

Yes. That map is similar to my AFR target map, except that I am holding off on the transition out of stoich for a bit longer on both the RPM and MAP scales, because I sometimes drive on the freeway with my 4.30 gearing (RPM) and I live in a very hilly area so I spend a lot of time at 70-80 kPa in normal everyday driving (MAP).



We're not ignoring your posts. We're just ignoring your feelings.

triple88a

iTrader: (2)

Man you guys are making me want to go back to NA engine now. I want to do a HHO hybrid but the turbo will cause problems such as making a pressurized bomb in a flashback incident that will probably make my intercooler look like a giant bong.

y8s

iTrader: (8)

big skim of the last two pages but thought I'd post this for joe's sake:

you will not generate enough EGT at cruise to raise any eyebrows. at 15:1 and higher AFR and relatively normal timing, I struggle to exceed 1200F.

triple88a

iTrader: (2)

In that case lean it out some more. What average temps are seen during cruise at say 50 mph 2.5k rpm?